This article is mentioned in:

Introduction

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) is commonly used to measure the expression levels of a selected gene of interest by quantifying the mRNA transcripts. RT-qPCR technology provides highly sensitive and accurate expression profiles (1,2). However, numerous parameters need to be controlled in order to obtain reliable quantitative expression measurements. These parameters include variations in the initial sample amount, RNA recovery, RNA integrity, efficiency of cDNA synthesis, and differences in the overall transcriptional activity of the tissue or cells analyzed (3). Therefore, reference genes are commonly used as internal controls. The reference gene is submitted to the same experimental protocol as the target gene in order to normalize qPCR data, and in order to reduce possible errors generated during the quantification of gene expression. The expression levels of the target gene are then normalized according to the values of the internal controls. Therefore, selecting a suitable reference gene is important for the correct interpretation of RT-qPCR analyses. Typically, reference genes satisfy the following criteria: They are highly expressed in the cells of interest, variability in the expression levels of the reference gene between samples is minimal, and the reference gene expression is not influenced by experimental conditions (2,4). However, it has recently become apparent that no single gene is constitutively expressed in all cell types and under all experimental conditions, implying that the expression stability of the intended control gene must be verified prior to each experiment. Therefore, the selection of a suitable reference gene is largely dependent on the experimental tissue samples.

At present, GAPDH, ACTB, and 18S rRNA are the most commonly used reference genes. However, previous studies have demonstrated that these reference genes are not always suitable, as they exhibit different behaviors in various types of cells and tissue (5,6), particularly in sheep. Due to their unique advantages (such as their similarity to humans in anatomy, physiology and wound repair mechanisms), sheep are widely used in medical experimental studies. Previous studies have successfully identified the most suitable reference genes for ovine nervous tissue, spleen, ileum, lung, mesenteric lymph node, pulmonary tissue, and blood (7–10). However, little is currently known regarding the most suitable reference genes for bone, specifically for the sheep mandibular condyle.

The present study aimed to identify a set of reference genes suitable for the normalization of sheep mandibular condyle RT-qPCR data. A total of 12 commonly-used reference genes belonging to various functional classes were selected, and their expression stability was determined in both normal and fractured sheep mandibular condyles. RefFinder was used to validate the reference genes.

Materials and methods

Animals

Nine healthy sheep (six male and three female; age, 1 year; weight, 30–40 kg) were included in the present study. The sheep were obtained from the Animal Experimental Center and housed at the School of Stomatology, both at The Fourth Military Medical University (Xi'an, China). Condylar fractures on the right side of the sheep mandibular were created as previously reported (11). Briefly, the zygomatic arch and the panniculus carnosus muscles at the surface of the capsule of the temporomandibular joint were exposed via a curved pre-auricular skin incision. A horizontal incision was subsequently made through the capsule at the condylar neck to open the inferior joint space. An oblique vertical osteotomy was made from the lateral pole of the condyle to the medial side of the condylar neck, using an ultrasound osteotome (Guilin Woodpecker Medical Instrument Co., Ltd., Guilin, China). The left side of the sheep mandibular served as a control. Three sheep were postoperatively sacrificed at the end of the second, fourth and twelfth week by dissecting the carotid artery (under a general anaesthetic) resulting in acute hemorrhagic death. The bones surrounding the fracture line on the right side, and the bones from the corresponding area on the left side were subsequently collected for further experimentation.

The present study was approved by the Ethics Committee of the School of Stomatology, The Fourth Military Medical University. All surgical procedures were conducted under satisfactory anesthesia (0.1 ml/kg xylazine hydrochloride; Military Veterinary Research Institute of Military Medical Science, Changchun, China). The sheep were cared for according to the guidelines set by the Laboratory Animal Research Centre of the Fourth Military Medical University.

Total RNA extraction and cDNA synthesis

Total RNA was extracted using the Total RNA kit I (Omega Bio-Tek, Inc., Norcross, GA, USA), according to the manufacturer's instructions. Contaminating genomic DNA was removed by on-column treatment of each sample with DNase I (Omega Bio-Tek, Inc., Norcross, GA, USA). The purity and quality of the extracted RNA was evaluated using a Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific, Inc., Waltham, MA, USA) at a wavelength of 260 nm. Total RNA (500 ng) was reverse transcribed in a final volume of 50 µl using PrimeScript® RT Reagent kit (Takara Bio, Inc., Otsu, Japan), according to the manufacturer's instructions. The cDNA was subsequently stored at −20°C.

Gene selection and primer design

A total of 12 genes belonging to various functional classes, which are frequently used as references genes in RT-qPCR studies, were selected for experimentation. The following genes were included in the present study: ACTB, YWHAZ, HPRT, TFRC, SDHA, B2M, PGK1, GAPDH, G6PD, RPL19, RPS24 and 18S. The primer sequences are specified in Table I (Augct, Beijing, China). The primer sequences for PGK1, SDHA, and G6PD were based on previous publications (7,9).

Table I Primer sequences of the 12 candidate reference genes.

Table I

Primer sequences of the 12 candidate reference genes.

Gene name	Primer sequences
RPL19	F: 5′ AGCCTGTGACTGTCCATTCC 3′ R: 5′ ACGTTACCTTCTCGGGCATT 3′
YWHAZ	F: 5′ AGACGGAAGGTGCTGAGAAA 3′ R: 5′ CGTTGGGGATCAAGAACTTT 3′
PGK1	F: 5′ ACTCCTTGCAGCCAGTTGCT 3′ R: 5′ AGCACAAGCCTTCTCCACTTCT 3′
HPRT	F: 5′ TTTATTCCTCATGGACTAATTATGGA 3′ R: 5′ CCACCCATCTCCTTCATCAC 3′
TFRC	F: 5′ TTCTGGGCAGACCTCAAATC 3′ R: 5′ CAGCTTCACGTGGGACATAA 3′
G6PD	F: 5′ TGACCTATGGCAACCGATACAA 3′ R: 5′ CCGCAAAAGACATCCAGGAT 3′
B2M	F: 5′ CTGTCGCTGTCTGGACTGG 3′ R: 5′ TTTGGCTTTCCATCTTCTGG 3′
ACTB	F: 5′ CCAACCGTGAGAAGATGACC 3′ R: 5′ CCAGAGGCGTACAGGGACAG 3′
SDHA	F: 5′ CATCCACTACATGACGGAGCA 3′ R: 5′ ATCTTGCCATCTTCAGTTCTGCTA 3′
RPS24	F: 5′ TTTGCCAGCACCAACGTTG 3′ R: 5′ AAGGAACGCAAGAACAGAATGAA 3′
18S	F: 5′ TGGTCGCTCGCTCCTCTCC 3′ R: 5′ CGCCTGCTGCCTTCCTTGG 3′
GAPDH	F: 5′ CCAGAGGCGGACTACTACG 3′ R: 5′ CCG AGA GAGCAACACAGG 3′

[i] F, forward; R, reverse.

RT-qPCR

All RT-qPCR reactions were performed in a 20 µl volume using SYBR® Premix Ex Taq™ II (Takara Bio, Inc.). The RT-qPCR was performed according to the instructions of the ABI 7500 real-time PCR system (Applied Biosystems Life Technologies, Foster City, CA, USA), which were as follows: Holding stage, 95°C for 3 min; 40 cycles of 95°C for 15 sec and 55°C for 34 sec; melt curve stage, 95°C for 15 secs, 60°C for 1 min, 95°C 30 sec and 60°C for 15 sec. A melt curve was included at the end of each RT-qPCR reaction.

Data analysis

RefFinder (http://www.leonxie.com/referencegene.php?type=reference) was used in order to validate the reference genes. RefFinder is a user-friendly web-based comprehensive tool developed for evaluating and screening reference genes from extensive experimental datasets. RefFinder integrates the following currently available computational algorithms: geNorm, NormFinder, BestKeeper, and the comparative ΔCt method, in order to compare and rank experimental candidate reference genes. Based on the rankings from each program, RefFinder assigns an appropriate weight to an individual gene, and calculates the geometric mean of these weights in order to generate a final overall ranking. In addition to analyses of the test and control groups, the expression data of the two groups were also combined to form a combined group.

Results

Expression levels of reference genes

All 12 reference genes expressed in the mandibular condyle of the sheep exhibited a single marked increase in the melt-curve. RPL19 and 18S exhibited the highest expression levels, with a comparative threshold (Ct) mean of 17.46 and 26.09, respectively, whereas B2M showed the lowest expression levels, with a Ct mean of 29.78. Ct refers to the cycle number at which the fluorescence intensity exceeds the set threshold. A high Ct value represents a low original template concentration, while a low Ct value represents a high original template concentration.

geNorm analysis

The stability values (M) of the reference genes were calculated using the geNorm algorithm. High M-values indicated high gene expression variability, whereas the most stable genes exhibited M-values <1.5. The M-values of the reference genes in the normal and fractured area of the sheep mandibular condyle as determined by geNorm, are shown in Table II and Fig. 1. In the control group, the most stable genes were RPL19 and ACTB, with an average M-value of 0.754. The second most stable gene was PGK1, with an average M-value of 0.948, whereas the least stable genes were SDHA and G6PD. However, in the test group, the most stable genes were S24 and B2M with an average M-value of 1.657, whereas the least stable genes were HPRT and YWHAZ. A comprehensive analysis of the combined group by geNorm identified the most stable genes to be RPL19 and ACTB, with an average M-value of 1.081. The second most stable gene in the combined group was PGK1, with an average M-value of 1.359. The remaining genes exhibited high M-values >1.5, with the least stable genes being SDHA and YWHAZ. Notably, SDHA and YWHAZ are considered conventional housekeeping genes in ovine blood (12).

Figure 1 Candidate reference genes ranked according to their expression stability, as determined by geNorm. (Y-axis, stability values of the reference genes).

Table II Expression stability of candidate reference genes, as determined by geNorm.

Table II

Expression stability of candidate reference genes, as determined by geNorm.

Gene name	M-value
	Control	Test	Combined
ACTB	0.745 (1)	1.778 (3)	1.081 (1)
RPL19	0.745 (1)	1.868 (4)	1.081 (1)
PGK1	0.948 (3)	2.110 (6)	1.359 (3)
18S	1.074 (4)	2.665 (9)	2.206 (7)
GAPDH	1.235 (5)	2.001 (5)	1.535 (4)
B2M	1.518 (6)	1.657 (1)	1.743 (5)
HPRT	1.713 (7)	3.182 (11)	2.774 (10)
S24	1.877 (8)	1.657 (1)	1.996 (6)
TFRC	2.037 (9)	2.480 (8)	2.387 (8)
YWHAZ	2.258 (10)	3.861 (12)	4.058 (12)
G6PD	2.473 (11)	2.350 (7)	2.553 (9)
SDHA	2.742 (12)	2.943 (10)	3.059 (11)

[i] Control, normal ovine condyle; test, fractured ovine condyle; combined, both normal and fractured ovine condyle. Numbers in brackets represent ranking of genes by expression stability.

NormFinder analysis

NormFinder ranks candidate genes according to their expression stability values (ϱ) based on the similarity of their expression profiles. Lower values are assigned to the most stable genes. The candidate gene expression stabilities, as determined by NormFinder, are shown in Table III and Fig. 2. In the control group, the expression stability analysis demonstrated that ACTB was the most stable gene, with a ϱ-value of 0.838. In the test group, PGK1 was identified as the most stable reference gene, with a ϱ-value of 1.209. When the expression data were combined (combined group), the results confirmed that RPL19 was the most stable gene, with a ϱ-value of 1.146.

Figure 2 Candidate reference genes ranked according to their expression stability, as determined by NormFinder. (Y-axis, stability values of the reference genes).

Table III Expression stability of candidate reference genes, as determined by NormFinder.

Table III

Expression stability of candidate reference genes, as determined by NormFinder.

Gene name	ϱ-value
	Control	Test	Combined
ACTB	0.838 (1)	1.642 (4)	1.252 (2)
HPRT	1.015 (2)	3.887 (11)	2.933 (10)
18S	1.073 (3)	2.851 (9)	2.266 (7)
RPL19	1.325 (4)	1.413 (2)	1.146 (1)
GAPDH	1.439 (5)	2.110 (7)	1.884 (6)
S24	1.675 (6)	2.086 (6)	2.468 (8)
PGK1	1.696 (7)	1.209 (1)	1.296 (3)
TFRC	1.937 (8)	1.928 (5)	1.723 (5)
B2M	2.064 (9)	2.233 (8)	1.679 (4)
YWHAZ	2.832 (10)	6.895 (12)	8.789 (12)
G6PD	3.060 (11)	1.602 (3)	2.563 (9)
SDHA	3.691 (12)	3.351 (10)	4.009 (11)

[i] Control, normal ovine condyle; test, fractured ovine condyle; combined, both normal and fractured ovine condyle. Numbers in brackets represent ranking of genes by expression stability.

BestKeeper analysis

Analysis of the reference gene expression levels was carried out using the BestKeeper software (Table IV and Fig. 3). In the control group, the results showed that ACTB was the most stable gene, with a stability value of 0.78. In the test group, PGK1 was identified as the most stable reference gene, with a stability value of 0.93. In the combined group, PGK1 was determined to be the most stably expressed reference gene with a stability value of 0.96, whereas YWAZH was the least stable reference gene.

Figure 3 Candidate reference genes ranked according to their expression stability, as determined by BestKeeper. (Y-axis, stability values of the reference genes).

Table IV Expression stability of candidate reference genes, as determined by BestKeeper.

Table IV

Expression stability of candidate reference genes, as determined by BestKeeper.

Gene name	Stability value
	Control	Test	Combined
ACTB	0.78 (1)	1.39 (2)	1.18 (2)
18S	0.87 (2)	2.51 (9)	1.88 (6)
PGK1	1.00 (3)	0.93 (1)	0.96 (1)
RPL19	1.08 (4)	1.66 (4)	1.46 (3)
GAPDH	1.42 (5)	1.81 (5)	1.80 (5)
S24	1.46 (6)	1.51 (3)	1.74 (4)
HPRT	1.50 (7)	4.16 (12)	2.69 (10)
B2M	1.98 (8)	2.28 (8)	2.16 (9)
TFRC	2.07 (9)	2.16 (7)	1.94 (7)
YWHAZ	2.23 (10)	3.23 (11)	4.93 (12)
G6PD	3.56 (11)	1.92 (6)	2.14 (8)
SDHA	4.08 (12)	2.51 (9)	2.98 (11)

[i] Control, normal ovine condyle; test, fractured ovine condyle; combined, both normal and fractured ovine condyle. Numbers in brackets represent ranking of genes by expression stability.

ΔCt analysis

The ΔCt method ranks a set of genes according to the repeatability of the gene expression differences among various samples (STDEV average). The results of the ΔCt analysis are shown in Table V and Fig. 4. In the control group, ACTB was the most stable reference gene followed by 18S, with stability values of 2.07 and 2.21, respectively. In the test group, G6PD and RPL19 were the highest stably expressed genes with stability values of 2.28 and 3.09, respectively. The combined results (combined group) identified RPL19 and ACTB as the most stable genes, with stability values of 3.02 and 3.06, respectively. YWHAZ and SDHA were the least stably expressed genes, with stability values of 9.05 and 4.88, respectively.

Figure 4 Candidate reference genes ranked according to their expression stability, as determined by the ΔCt method. (Y-axis, stability values of the reference genes).

Table V Expression stability of candidate reference genes, as determined by the ΔCt method.

Table V

Expression stability of candidate reference genes, as determined by the ΔCt method.

Gene name	Stability value
	Control	Test	Combined
ACTB	2.07 (1)	3.11 (2)	3.06 (2)
18S	2.21 (2)	3.91 (9)	3.67 (8)
RPL19	2.24 (3)	3.09 (1)	3.02 (1)
HPRT	2.30 (4)	4.65 (11)	4.18 (10)
GAPDH	2.38 (5)	3.38 (6)	3.37 (4)
PGK1	2.47 (6)	3.12 (3)	3.16 (3)
S24	2.63 (7)	3.35 (5)	3.64 (7)
B2M	2.77 (8)	3.42 (8)	3.42 (5)
TFRC	2.79 (9)	3.41 (7)	3.49 (6)
YWHAZ	3.40 (10)	7.25 (12)	9.05 (12)
G6PD	11.00 (11)	2.28 (4)	3.77 (9)
SDHA	12.00 (12)	4.36 (10)	4.88 (11)

[i] Control, normal ovine condyle; test, fractured ovine condyle; combined, normal and fractured ovine condyle. Numbers in brackets represent ranking of genes by expression stability.

Comprehensive analysis

The combined results obtained from geNorm, BestKeeper, the ΔCt method, and NormFinder were analyzed comprehensively using RefFinder (Table VI and Fig. 5). In the control group, RefFinder identified ACTB as the most stably expressed gene, with a stability value of 1.00. In the test group, PGK1 was the most stable reference gene with a stability value of 2.06. The combined results of the two groups (combined group) identified RPL19 as the most stably expressed reference gene with a stability value of 1.32, whereas YWAZH was the least stable reference gene. Candidate reference genes of the control group, test group and combined group, ranked using five different methods, were compared in Figs. 6Figure 7–8, respectively. Though each algorithm employs a different method to determine the most suitable reference gene, the results obtained from all of the methods exhibited a similar trend.

Figure 5 Candidate reference genes ranked, as determined by comprehensive analysis. (Y-axis, stability values of the reference genes).

Figure 6 Candidate reference genes of the control group ranked using five different methods. (Y-axis, stability values of the reference genes).

Figure 7 Candidate reference genes of the test group ranked using five different methods. (Y-axis, stability values of the reference genes).

Figure 8 Candidate reference genes of the combined group ranked using five different methods. (Y-axis, stability values of the reference genes).

Table VI Expression stability of candidate reference genes, as determined by RefFinder.

Table VI

Expression stability of candidate reference genes, as determined by RefFinder.

Gene name	Stability value
	Control	Test	Combined
ACTB	1.00 (1)	2.63 (3)	1.68 (2)
RPL19	2.63 (2)	2.38 (2)	1.32 (1)
18S	2.63 (3)	9.00 (9)	6.96 (8)
PGK1	4.41 (4)	2.06 (1)	2.28 (3)
HPRT	4.45 (5)	11.24 (11)	10.00 (10)
GAPDH	5.00 (6)	5.69 (7)	4.68 (4)
S24	6.70 (7)	3.08 (4)	6.05 (6)
B2M	7.67 (8)	4.76 (6)	5.48 (5)
TFRC	8.74 (9)	6.65 (8)	6.40 (7)
YWHAZ	10.00 (10)	11.74 (12)	2.00 (12)
G6PD	11.00 (11)	4.74 (5)	8.74 (9)
SDHA	12.00 (12)	10.00 (10)	11.00 (11)

[i] Control, normal ovine condyle; test, fractured ovine condyle; combined, both normal and fractured ovine condyle. Numbers in brackets represent ranking of genes by expression stability.

Discussion

Due to its high specificity, sensitivity, efficiency and repetitiveness, RT-qPCR has been widely used to measure the expression levels of target genes. The advantages of RT-qPCR also includes rapid detection, simple design, and low cost. However, the primary limitation of relative quantification is that a reference gene must be included as an internal control. Numerous reference genes, such as GAPDH, ACTB, 18S, and G6PD, have previously been identified. However, to date, no reference gene has been shown to be constitutively expressed in all cell types and under all experimental conditions. Therefore, evaluating the expression stability of candidate reference genes and identifying the most stable gene is important in order to ensure the authenticity and reliability of a given study. Numerous reference genes have been identified as suitable for various ovine tissue samples. SDHA and YWHAZ are considered to be the most suitable internal controls for RT-qPCR analysis of ovine blood (12), ACTB for the cerebrum and spleen, and GAPDH for the mesenteric lymph node (7). However, little is currently known regarding the most suitable reference gene for the bone, particularly for the sheep mandibular condyle. Sheep have an important role in the study of human bone diseases, particularly studies associated with temporomandibular diseases (13,14). Therefore, the identification of a reference gene suitable for the normalization of sheep mandibular condyle RT-qPCR analyses is necessary.

In the present study, the expression stability of 12 housekeeping genes were examined in both normal and fractured sheep mandibular condyles using the following algorithms: geNorm, NormFinder, BestKeeper, and the ΔCt method. geNorm (15) calculates the expression ratio of two ideal internal control genes in two identical samples, regardless of the experimental conditions or cell type. For each control gene, the pairwise variation is determined as the standard deviation of the logarithmically transformed expression ratios, and the internal control gene-stability value (M) is determined as the average pairwise variation of a particular gene with all other control genes. NormFinder (3) ranks the candidate genes according to their expression stability values (ϱ) based on the similarity of their expression profiles. BestKeeper (16) identifies the most suitable reference genes using a pairwise correlation analysis of all pairs of candidate genes, and calculates the geometric mean of the most suitable genes. The ΔCt (17) method ranks a set of genes according to the repeatability of the gene expression differences among various samples (average of STDEV). Although each algorithm employs a different method to determine the most suitable reference gene, in the present study the results obtained from all five methods exhibited a similar trend (Fig. 6Figure 7–8). All methods determined that RPL19, ACTB, and PGK1 were the most stably expressed reference genes in the sheep mandibular condyle, regardless of whether the candidate genes were analyzed for the normal group, the fracture group, or the combined group. The results obtained from the various algorithms were comprehensively analyzed using RefFinder. Based on the rankings obtained from each algorithm, RefFinder assigns an appropriate weight to an individual gene, and calculates the geometric mean of these weights, in order to obtain an overall final ranking. The results demonstrated that ACTB and RPL19 were the most stable reference genes expressed in the normal sheep mandibular condyle, PGK1 and RPL19 were the most stable reference genes expressed in the fractured sheep mandibular condyle, whereas RPL19 and ACTB were the most stably expressed reference genes in both the normal and fractured sheep mandibular condyle (combined group). The results of the present study identified RPL19 as the most suitable reference gene for studies associated with the sheep mandibular condyle. In addition, ACTB and PGK1 were shown to be suitable alternatives. These results differ from previous reports of sheep reference gene suitability, which included the identification of SDHA and YWHAZ as the most stable reference genes for ovine blood (12), ACTB for the cerebrum and spleen, and GAPDH for the mesenteric lymph node (7). In conclusion, the results of the present study suggested that if these previously reported genes were to be used as reference genes in studies associated with the sheep mandibular condyle, the results of such a study may be negatively affected.

Acknowledgments

The present study was supported by a grant from the National Natural Science Foundation of China (grant no. 81271168).

References